Source code for the Rhode Island Archival and Manuscript Collections Online (RIAMCO) website.

We are currently using Ruby 2.7.1, Rails 6.0.2, MySQL, and Solr 7.

brew install ruby-install
brew install chruby
ruby-install ruby 2.7.1
source /usr/local/opt/chruby/share/chruby/chruby.sh
chruby 2.7.1
gem install bundle

See ./solr_conf/solr_create.sh for instructions on how to create the Solr core required by this project.

Update the values in .env_sample to match the URLs where Solr is running in your environment.

git https://github.com/Brown-University-Library/riamco.git
cd riamco
bundle install
source .env_sample
bundle exec rake db:migrate
bundle exec rails server

bundle exec rake riamco:import_eads[/path/to/riamco/sampledata/*.xml]

You can use rake task parse_eads if you just want to parse an EAD and see the result in your Terminal.

The code to convert the finding aids from XML to Solr documents is in ./app/models/ead.rb and ./app/models/ead_import.rb.

If you are interested in indexing the content of the PDF files indicated in an EAD (in addition to the EAD itself) there are a few of extra steps required.

Download the Tika Server from the Apache Tika website and run it (leave it running)

curl https://apache.mirrors.tds.net/tika/tika-server-1.22.jar > tika-server-1.22.jar
java -jar tika-server-1.22.jar

Run the following Rake task to scan a particular EAD (by EAD ID) and index the PDF files indicated on it.

bundle exec rake riamco:ft_index_ead[US-RPB-ms2018.010]

The code to extract the content of the PDF files and index it in Solr is in ./app/models/full_text_import.rb.

Most of the search logic is in ./app/controllers/search_controller.rb and ./app/models/search.rb.

The code to view individual finding aids is in ./app/controllers/ead_controller.rb and relies heavily on the XSLT files under ./xslt/.

See this document.